Toxic effects of brake wear particles on epithelial lung cells in vitro

Abstract

Background: Fine particulate matter originating from traffic correlates with increased morbidity and mortality. An important source of traffic particles is brake wear of cars which contributes up to 20% of the total traffic emissions. The aim of this study was to evaluate potential toxicological effects of human epithelial lung cells exposed to freshly generated brake wear particles.

Results: An exposure box was mounted around a car's braking system. Lung cells cultured at the air-liquid interface were then exposed to particles emitted from two typical braking behaviours ("full stop" and "normal deceleration"). The particle size distribution as well as the brake emission components like metals and carbons was measured on-line, and the particles deposited on grids for transmission electron microscopy were counted. The tight junction arrangement was observed by laser scanning microscopy. Cellular responses were assessed by measurement of lactate dehydrogenase (cytotoxicity), by investigating the production of reactive oxidative species and the release of the pro-inflammatory mediator interleukin-8. The tight junction protein occludin density decreased significantly (p < 0.05) with increasing concentrations of metals on the particles (iron, copper and manganese, which were all strongly correlated with each other). Occludin was also negatively correlated with the intensity of reactive oxidative species. The concentrations of interleukin-8 were significantly correlated with increasing organic carbon concentrations. No correlation was observed between occludin and interleukin-8, nor between reactive oxidative species and interleukin-8.

Conclusion: These findings suggest that the metals on brake wear particles damage tight junctions with a mechanism involving oxidative stress. Brake wear particles also increase pro-inflammatory responses. However, this might be due to another mechanism than via oxidative stress.

Figure 1

Characterisation of brake wear particles. A) shows a typical contribution of measured brake particles from a braking simulation with 8 repetitions, „normal deceleration“. B) shows the particle number, mass and surface concentrations in the air of different braking behaviours. Values represent the sum of all concentrations measured during one run ("sum under the curve"). Such a representation enables a better comparability among runs with different exposure times. SD represents the standard deviation. §p < 0.05, §§p < 0.01 represent a statistically significant influence of the fixed factors. + stands for p < 0.05, ++ for p < 0.01 in pairwise Bonferroni t-tests of designated braking conditions in comparison with the "full stop 8x". C) and D) represent identical illustrations with measured metals and carbon components. Displayed are the means and the standard deviations from 2 ("full stop 4x") or 4 (all the other conditions) experiments. Standard deviations for the elemental carbon could not be calculated for all bars because of values below the detection limit.

Figure 2

Number of deposited particles and correlation of the number of deposited particles with total particle number. A) Particle number/cm² was calculated by counting the individual particles deposited on a TEM grid. B) Total particle number (sum under the curve) was measured online during one exposure and correlated with the number of deposited particles.

Figure 3

Cytotoxicity in exposed A549 cells compared with control cultures. LDH levels were determined as a marker of cell death and the values did not change significantly after exposure to brake wear particles. Each value represents the mean and standard deviations from 3 individual exposure experiments.

Figure 4

Production of reactive oxygen species in exposed and non exposed cells. Picture A) shows cells that were not exposed (negative). Picture B) stands for the positive control (treated with TBHP). Cells on pictures C) and D) were exposed to brake wear particles ("normal deceleration 8x" and "full stop 8x"). The Intensity of ROS from A549 cells exposed to brake particles was quantified with the software ITEM (E). Displayed are the means and the standard deviations from 2 ("full stop 4x") or 4 (all the other conditions) experiments.

Figure 5

Influence of brake wear particles on the tight junction protein occludin. A) Non exposed cells. B) Cells exposed to a „full stop“ with 8 repetitions. The graph on the bottom shows the quantifications of occludin (B). Each value represents the median of the 25%- and the 75%-quantile from 2 ("full stop 4x" and "no stop run") or 4 (all the other conditions) experiments.

Figure 6

Concentrations of IL-8 released from A549 cells exposed to different braking behaviours. Means and standard deviations from 2 ("full stop 4x" and „no stop run“) or 4 (all the other conditions) experiments. A positive control was generated with TNF-α stimulation.

Figure 7

Selection of correlating factors. A-C shows the metals iron, copper and manganese correlating with occludin. D and E are the correlations of total carbon and organic carbon with measured IL-8 concentrations. The spotted line shows the linear correlation of the medians.